Chapter 29 The Theory of Special Status Pictures" and "Imagining" Gilbert Ryle

Let us first consider some implications of the other doctrine, that in visualising I am, in a nearly ordinary sense of the verb, seeing a picture with a special status. It is part of this doctrine that the picture that I see is not, as snapshots are, in front of my face; on the contrary, it has to be not in physical space, but in a space of another kind.

The pictured smile is not, then, a physical phenomenon, i.e. a real contortion of the doll's face; nor yet is it a non-physical phenomenon observed by the child taking place in a field quite detached from her perambulator and her nursery. There is not a smile at all, and there is not an effigy of a smile either. There is only a child fancying that she sees her doll smiling. So, though she is really picturing her doll smiling, she is not looking at a picture of a smile; and though I am fancying that I see rabbits coming out of the hat, I am not seeing real phantasms of rabbits coming out of real phantasms of hats.

Take another case. I start to write down a long and unfamiliar word and after a syllable or two, I find that I am not sure how the word should go on. I then, perhaps, imagine myself consulting a dictionary and in some cases I can then 'see' how the last three syllables are printed. In this sort of case it is tempting to say that I am really seeing a picture of a printed word, only the picture is 'in my head', or 'in my mind', since reading off the letters of the word that I 'see' feels rather like reading off the letters from a dictionary-item, or a photograph of such an item, which I really do see. But in another case, I start writing the word and I 'see' the next syllable or two on the page on which I am writing and in the place where I am to write them. I feel rather as if I were merely inking in a word-shadow lying across the page. Yet here it is impossible to say that I am having a peep at a picture or ghost of a word in a queer space other than physical space, for what I 'see' is on my page just to the right of my nib. Again we must say that though I picture the word in a certain place, printed in a certain type, or written in a certain handwriting, and though I can read off the spelling of the word from the way I picture it as printed or written, yet there exists no picture, shadow or ghost of the word and I see no picture, shadow or ghost of it. I seem to see the word on the page itself, and the more vividly and sustainedly I seem to see it, the more easily can I transcribe what I seem to see on to my paper with my pen.

Hume notoriously thought that there exist both 'impressions' and 'ideas', that is, both sensations and images; and he looked in vain for a clear boundary between the two sorts of 'perceptions'.

Hume's attempt to distinguish between ideas and impressions by saying that the latter tend to be more lively than the former was one of two bad mistakes. Suppose, first, that 'lively' means 'vivid'. A person may picture vividly, but he cannot see vividly. One 'idea' may be more vivid than another 'idea', but impressions cannot be described as vivid at all, just as one doll can be more lifelike than another, but a baby cannot be lifelike or unlifelike. To say that the difference between babies and dolls is that babies are more lifelike than dolls is an obvious absurdity. One actor may be more convincing than another actor; but a person who is not acting is neither convincing nor unconvinc­ing, and cannot therefore be described as more convincing than an actor. Alterna­tively, if Hume was using 'vivid' to mean not lifelike' but 'intense', 'acute' or 'strong', then he was mistaken in the other direction; since, while sensations can be compared with other sensations as relatively intense, acute or strong, they cannot be so compared with images.

Similarly, there are not two species of murderers, those who murder people, and those who act the parts of murderers on the stage; for these last are not murderers at all. They do not commit murders which have the elusive attribute of being shams; they pretend to commit ordinary murders, and pretending to murder entails, not murdering, but seeming to murder. As mock-murders are not murders, so imagined sights and sounds are not sights or sounds. They are not, therefore, dim sights, or faint sounds. And they are not private sights or sounds either. There is no answer to the spurious question, Where have you deposited the victim of your mock-murder?' since there was no victim. There is no answer to the spurious question, Where do the objects reside that we fancy we see?' since there are no such objects.

It will be asked, ^zHow can a person seem to hear a tune running in his head, unless there is a tune to hear7^, Part of the answer is easy, namely that he would not be seeming

"The Theory of Special Status Pictures" and "Imagining" 205 to hear, or fancying that he heard, a tune, if he were really hearing one, any more than the actor would be simulating murder, if he were really murdering someone. But there is more to be said than this. The question, 'How can a person seem to hear a tune, when there is no tune to be heard?' has the form of a 'wires and pulleys' question. It suggests that there exists a mechanical or para-mechanical problem, (like those that are properly asked about conjuring-tricks and automatic telephones), and that we need to have described to us the hidden workings that constitute what a person does, when he fancies himself listening to a tune.

Let us consider another sort of imaging. Sometimes, when someone mentions a blacksmith's forge, I find myself instantaneously back in my childhood, visiting a local smithy. I can vividly 'see' the glowing red horseshoe on the anvil, fairly vividly 'hear' the hammer ringing on the shoe and less vividly 'smell' the singed hoof. How should we describe this 'smelling in the mind's nose"? Ordinary language provides us with no means of saying that Γ am smelling a likeness' of a singed hoof.

None the less, I may certainly say that I vividly 'smell' the singed hoof, or that its smell comes back to me vividly, and the use of this adverb shows by itself that I know that I am not smelling, but only 'smelling'. Smells are not vivid, faithful or lifelike; they are only more or less strong. Only 'smells' can be vivid, and correspondingly they

cannot be more or less strong, though I can seem to be getting a more or less strong smell. However vividly I may be 'smelling' the smithy, the smell of lavender in my room, however faint, is in no degree drowned. There is no competition between a smell and a 'smell', as there can be a competition between the smell of onions and the smell of lavender.

If a person who has recently been in a burning house reports that he can still 'smell' the smoke, he does not think that the house in which he reports it is itself on fire. However vividly he 'smells' the smoke, he knows that he smells none; at least, he realises this, if he is in his right mind, and if he does not realise it, he will say not that the 'smell' is vivid, but, erroneously, that the smell is strong. But if the theory were true that to 'smell' smoke were really to smell a likeness of smoke, he could have no. way of distinguishing between 'smelling' and smelling, corresponding to the familiar ways in which we distinguish between looking at faces and looking at likenesses of them, or between hearing voices and hearing recordings of voices.

There are usually ocular ways of distinguishing between things and snapshots or effigies of them; a picture is flat, has edges and perhaps a frame; it can be turned round and turned upside down, crumpled and tom. Even an echo, or a recording, of a voice can be distinguished, if not audibly, at least by certain mechanical criteria from the voice itself. But no such discriminations can be made between a smell and a copy of a smell, a taste and a likeness of a taste, a tickle and a dummy-tickle; indeed, it makes no sense to apply words like 'copy', likeness' and 'dummy' to smells, tastes and feelings. Conse­quently we have no temptation to say that a person who 'smells' the smithy is really smelling a facsimile or likeness of anything. He seems to smell, or he fancies he smells, something, but there is no way of talking as if there existed an internal smell replica, or smell facsimile, or smell echo. In this case, therefore, it is clear that to 'smell' entails not smelling and therefore that imaging is not perceiving a likeness, since it is not perceiv­ing at all.

Why, then, is it tempting and natural to misdescribe 'seeing things' as the seeing of pictures of things? It is not because 'pictures' denotes a genus of which snapshots are one spedes and mental pictures are another, since 'mental pictures' no more denotes pictures than 'mock-murders' denotes murders. On the contrary, we speak of 'seeing' as if it were a seeing of pictures, because the familiar experience of seeing snapshots of things and persons so often induces the 'seeing' of those things and persons. This is what snapshots are for. When a visible likeness of a person is in front of my nose, I often seem to be seeing the person himself in front of my nose, though he is not there and may be long since dead. I should not keep the portrait if it did not perform this function. Or when I hear a recording of a friend's voice, I fancy I hear him singing or speaking in the room, though he is miles away. The genus is seeming to perceive, and of this genus one very familiar spedes is that of seeming to see something, when looking at an ordinary snapshot of it. Seeming to see, when no physical likeness is before the nose, is another spedes. Imaging is not having shadowy pictures before some shadow-organ called 'the mind's eye'; but having paper pictures before the eyes in one's face is a familiar stimulus to imaging.

An oil painting of a friend is described as lifelike, if it makes me seem to see the friend in great clarity and detail, when I am not actually seeing him. A mere cartoon may be lifelike without being at all similar to a lifelike oil painting of the same person. For a picture to be lifelike it is not necessary or suffident that it should be an accurate replica of the contours or colouring of the subject's face. So when I vividly 'see' a face, this does not entail my seeing an accurate replica, since I might see an accurate replica without being helped to 'see' the face vividly and vice versa. But finding a picture of a person

"The Theory of Special Status Pictures" and "Imagining" 207 lifelike or 'speaking' entails being helped to seem to see the person, since that is what 'lifelike' and 'speaking' mean.

People have tended to describe 'seeing' as a seeing of genuine but ghostly likenesses, because they wanted to explain vividness or Iifelikeness in terms of similarity, as if, for me vividly to 'see' Helvellyn, I must be actually seeing something else very similar to Helvellyn. But this is erroneous. Seeing replicas, however accurate, need not result in 'seeing' vividly, and the speakingness of a physical likeness has to be described, not in terms of similarity, but in terms of the vividness of the 'seeing' which it induces.

In short, there are no such objects as mental pictures, and if there were such objects, seeing them would still not be the same thing as seeming to see faces or mountains. We do picture or visualise faces and mountains, just as we do, more rarely, 'smell' singed hoofs, but picturing a face or a mountain is not having before us a picture of the face or mountain, it is something that having a physical likeness in front of one's nose com­monly helps us to do, though we can and often do do it without any such promptings. Dreaming, again, is not being present at a private cinematograph show; on the con­trary, witnessing a public cinematograph show is one way of inducing a certain sort of dreaming. The spectator there is seeing a variously illuminated sheet of linen, but he is 'seeing' rolling prairies. So it would invert the true state of affairs to say that the dreamer is regarding a variously illuminated sheet of 'mental' linen; for there is no mental linen, and if there were, seeing it variously illuminated would not be dreaming that one was galloping over the prairies.

The tendency to describe visualising as seeing genuine, but internal, likenesses, reinforces and is reinforced by the Sense Datum Theory. Many holders of this theory, supposing, erroneously, that in 'seeing' I am seeing a peculiar paper-less snapshot, though one which, oddly, cannot be turned upside down, think that a fortiori in seeing proper I am seeing a peculiar non-physical colour expanse. And supposing, errone­ously, that having a visual sensation is descrying a flat patchwork of colours spread out in 'a private space', they And it all the easier to say that in imaging we are scanning a more ghostly patchwork of colours hung up in the same gallery with that original patchwork of colours. As in my study there may be both a person and a shadow or a portrait of that person, so in my private sight-gallery there might be both sense data and reproductions of sense data. My objections to the interpretation of picturing as picture-seeing do not in themselves demolish the Sense Datum Theory of sensations; but they do demolish, I hope, the ancillary theory that picturing is looking at reproduc­tions of sense data. And if I am right in saying that having a visual sensation is wrongly described as some sort of observing of a patchwork of colours, since the concept of sensation is different from the concept of observing, it will follow, as can be established on other grounds, that imaging is not only not any sort of observing of anything; it is also not having a sensation of a special sort. Seeming to hear a very loud noise is not being in any degree deafened, nor is seeming to see a very bright light being in any degree dazzled. So far are ideas from being impressions of a special sort, that to describe something as an idea, in this sense, is to deny that an impression is being had.

It will probably be asked, What then is it for a person to fancy that he sees or smells something? How can he seem to hear a tune that he does not really hear? And, in particular, how can a person fail to be aware that he is only seeming to hear or see, as the dipsomaniac certainly fails? In what precise respects is 'seeing' so like seeing that the victim often cannot, with the best will and the best wits, tell which he is doing?* Now if we divest these questions of associations with any 'wires and pulleys' questions, we can see that they are simply questions about the concept of imagining or make- believe, a concept of which I have so far said nothing positive. I have said nothing about it so far, because it seemed necessary to begin by vaccinating ourselves against the theory, often tacitly assumed, that imagining is to be described as the seeing of pictures with a special status.

But I hope I have now shown that what people commonly describe as having a mental picture of Helvellyn' or having Helvellyn before the mind's eye' is actually a special case of imagining, namely imagining that we see Helvellyn in front of our noses, and that having a tune running in one's head is imagining that one has the tune being played in one's hearing, maybe in a concert-hall. If successful, then I have also shown that the notion that a mind is a 'place', where mental pictures are seen and reproductions of voices and tunes are heard, is also wrong.

There are hosts of widely divergent sorts of behaviour in the conduct of which we should ordinarily and correctly be described as imaginative. The mendacious witness in the witness-box, the inventor thinking out a new machine, the constructor of a ro­mance, the child playing bears, and Henry Irving are all exercising the imaginations; but so, too, are the judge listening to the lies of the witness, the colleague giving his opinion on the new invention, the novel reader, the nurse who refrains from admon­ishing the bears' for their subhuman noises, the dramatic critic and the theatre-goers. Nor do we say that they are all exercising their imaginations because we think that, embedded in a variety of often widely different operations, there is one common nuclear operation which all alike are performing, any more than we think that what makes two men both farmers is some nuclear operation which both do in exactly the same way. Just as ploughing is one farming job and tree-spraying is another farming job, so inventing a new machine is one way of being imaginative and playing bears is another. No one thinks that there exists a nuclear farming operation by the execution of which alone a man is entitled to be called 'a farmer⁷; but the concepts wielded in theories of knowledge are apt to be less generously treated. It is often assumed that there does exist one nuclear operation in which imagination proper consists; it is assumed, that is, that the judge following the witness's mendacities, and the child playing bears, are both exercising their imaginations only if they are both executing some specifically identical ingredient operation. This supposed nuclear operation is often supposed to be that of seeing things in the mind's eye, hearing things in one's head and so on, i.e. some piece of fancied perceiving. Of course, it is not denied that the child is doing lots of other things as well; he roars, he pads around the floor, he gnashes his teeth and he pretends to sleep in what he pretends is a cave. But, according to this view, only if he sees pictures in his mind's eye of his furry paws, his snowbound den and so on, is he imagining anything. His noises and antics may be a help to his picturing, or they may be special effects of it, but it is not in making these noises, or performing these antics, that he is exercising his imagination, but only in his 'seeing', bearing', 'smelling', 'tast­ing' and 'feeling' things which are not there to be perceived. And the corresponding things will be true of the attentive, if sceptical, judge.

Put as bluntly as this, the doctrine is patently absurd. Most of the things for which we ordinarily describe children as imaginative are ruled out in favour of a limited number of operations the occurrence and qualities of which it is difficult to ascertain, especially from relatively inarticulate children. We see and hear them play, but we do not see or hear them 'seeing' or bearing' things. We read what Conan Doyle wrote, but we do not get a view of what he saw in his mind's eye. So, on this theory, we cannot easily tell whether children, actors or novelists are imaginative or not, though the word 'imagination' came to be wielded in theories of knowledge just because we all know how to wield it in our everyday descriptions of children, actors and novelists.

'The Theory of Special Stahis Pictures" and "Imagining" 209

There is no special Faculty of Imagination, occupying itself single-mindedly in fan­cied viewings and hearings. On the contrary, 'seeing' things is one exerdse of imagina­tion, growling somewhat like a bear is another; smelling things in the mind's nose is an uncommon act of fancy, malingering is a very common one, and so forth. Perhaps the chief motive from which many theorists have limited the exercises of imagination to the special class of fancied perceptions is that they have supposed that, since the mind is officially tri-partitioned into the Three Estates of Cognitioa Volition and Emotioa and since imagination was bom into the first, it must therefore be excluded from the others. Cognitive malpractices are notoriously due to the pranks of undisciplined Imaginatioa and some cognitive successes are in debt to its primmer activities. So, being an (erratic) Squire of Reasoa it cannot serve the other masters. But we need not pause to discuss this feudal allegory. Indeed, if we are asked whether imagining is a cognitive or a noncognitive activity, our proper policy is to ignore the question. 'Cognitive' belongs to the vocabulary of examination papers.